Comments on the ICMJE’s Proposals for Sharing Clinical Trial Data

Displaying 316 - 320 of 320 comments
  • David Raymer
    Washington University / Barnes Jewish Hospital
    Role(s):
    • Researcher
    • Clinician
    • Clinical trial participant
    Date Submitted: Wednesday, January 20, 2016 - 13:03

    Requirement To Share Data
    • I agree with this general approach

    6 Month Time Frame
    • I agree with this general approach

    Require a Data Sharing Plan
    • I agree with this general approach

    Providing Credit
    • I agree with this general approach

    Position:
    Comments:
  • Walter Palmas MD
    Columbia University
    Role(s):
    • Researcher
    • Clinician
    • Clinical trialist
    Date Submitted: Wednesday, January 20, 2016 - 12:07

    Requirement To Share Data
    • I agree with this general approach

    6 Month Time Frame
    • I agree with this general approach
    Comments: This should be extended to 12 - 18 months. Six months may not allow important follow-up manuscripts to be drafted.

    Require a Data Sharing Plan
    • I agree with this general approach
    Comments: There should be a requirement to share the approved consent (and assent) forms, too. An ethical approach to recruitment should include Consent Forms that can withstand third party scrutiny.

    Providing Credit
    • I agree with this general approach
    Comments: Study investigators should be offered a collaboration, with authorship, whenever feasible. This is particularly important when the new publication is based on data from a single study.

    Other Comments: The Committee should be commended for taking this on.

    Position:
    Comments:
  • Rachel Jewkes
    South African Medical Research Council
    Role(s): Researcher
    Date Submitted: Wednesday, January 20, 2016 - 11:43

    Requirement To Share Data
    Comments: Open data access is fundamentally unfair and will undermine research and knowledge production in developing countries. It may be appropriate for certain types of clinical research, but it is not appropriate for LMIC social epidemiology. This process will establish a competition between those who collect data and want to be able to maximally use it for their own intellectual ends (and those of their team), and those who wait for opportunities to benefit from the hard work of the former (the data vultures). Why is this seen as being desirable? In social epidemiology, I have observed that papers based on further analysis of data conducted by people who are disconnected from the field at best produce science which is mediocre and contributes little to advancing knowledge, but at worst produces misleading publications with misinterpretation of findings due to a lack of detailed familiarity with the issues. This isn't good science. I will not be able to train PhDs and post-docs in a climate of data competition.

    6 Month Time Frame
    Comments: Secondary analysis is a great way to make a career - no dirty hands, no staff management, no risk exposure in the field etc. Its not good for science. It advantages people disconnected from the field and issues and who are not placed to conduct proper engagement with national stakeholders around research findings of the type which is really vital for changing health policy and ultimately people's lives. In South Africa my staff and I spend hours managing our trials, training and supporting staff and undertake very extensive national research uptake. What all of this means is that we have far less time for data analysis and writing that most northern clinical trialists (or those doing secondary analysis). We need much longer to justice to a dataset and need to be able to do so without the worry (and threat) of being pipped to the post by a researchers who have no prior connection with the data. It will be to the detriment of science if our ability to develop the capacity (and careers) of southern researchers is compromised by this.

    Require a Data Sharing Plan
    Comments: I believe that people who want to analyse data should collect it.

    Providing Credit
    Comments: The only credit which is worth anything is co-authorship after a process of collaborative writing, but those generating data would have to have the right to decline collaboration and access.

    Other Comments: This whole data sharing movement has been driven by concerns of the global north. There really has been very little scope for the voices and views of the south in the process. Given the history of colonialism and neo-colonialism I cannot say this is unprecedented but it is profoundly disappointing in the twenty-first century. Its very easy to draw parallels between this and the post-colonial pattern of economic exploitation where basic resources are mined and harvested in southern countries for processing and wealth creation in the global north. I would suggest that a much more constructive approach is to encourage data sharing rather than coercive data sharing and to recognise the value of, and need for, diversity.

    Position:
    Comments:
  • Steven Bellan
    University of Texas at Austin
    Role(s): Researcher
    Date Submitted: Wednesday, January 20, 2016 - 10:56

    Requirement To Share Data
    • I agree with this general approach
    Comments: That clinical trial data (much of which is funded publicly) does not see the light of day is appalling given that further analyses can both uncover mistakes, generate new hypotheses, and more generally provide greater insight into factors underlying human health. I fully support this requirement. I think that this should be the default--deidentified data release within 6 months.

    6 Month Time Frame
    • I agree with this general approach
    Comments: Given the pressure to publish, scientists often want to be able to milk more out of their data before others can do so. I think that there should be some room to request an extension past this 6 month period, say up to 18 months. However, I would suggest that an extension require a formal extension request submission justifying why the extension is needed and proposing a timeline of the planned research/activities to happen in the 6-18 month period that necessitate the extension. Journals can then vet these extensions on a case-by-case basis. If authors repeatedly fail to produce the results motivating an 18-month extension, they should be less likely to receive future extensions.

    Require a Data Sharing Plan
    • I agree with this general approach

    Providing Credit
    • I agree with this general approach
    Comments: Data-set citations could easily be developed to show how many articles rely on published data, these could be weighted by impact factor or the like to show how influential publication of data availability is.

    Other Comments: I also think that these requirements should be applied to some extent to observational data--perhaps with less stringent requirements given that such data are often far more comprehensive since they can be collected through passive surveillance or other larger data-collection systems than possible in a clinical trial. Still, the specific data set (i.e. variables, time period, etc...) that forms part of a published analysis could be released by a given deadline.

    Position:
    Comments:
  • James Goedert, MD
    National Cancer Institute, NIH
    Role(s): Researcher
    Date Submitted: Wednesday, January 20, 2016 - 10:46

    Requirement To Share Data
    • I agree with this general approach

    6 Month Time Frame
    • I agree with this general approach

    Require a Data Sharing Plan
    • I agree with this general approach

    Providing Credit
    • I agree with this general approach

    Position:
    Comments:

Pages